As the background value is an important factor affecting the precision of grey system model, a method for optimizing the background value in the grey Verhulst model is proposed. Based on the Logistic function structure of time response formula in the grey Verhulst model and the geometric meaning of background value formula, the mean value theorem of integral is used to study the numerical relationship between the background value and the growth coefficient. New parameters are evaluated by using least square method, and the estimated value of original parameters is calculated by using equations. The model with the new background value satisfies the unbiased and least error. The examples show that the precision of simulation of the optimized model is obviously higher and the efficiency and feasibility of the optimized model are validated.